AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

ECG-QA: A Comprehensive Question Answering Dataset Combined With Electrocardiogram

Jungwoo Oh, Gyubok Lee, Seongsu Bae, Joon-myoung Kwon, Edward Choi

NeurIPS 2023 · Large language models › Retrieval and knowledge · approach: Domain-specific QA datasets · 3 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use pre-trained multimodal model (M3AE) instead of Fusion TransformerECG-QA baseline · ECG-QA test, S-ChooseEM accuracy: 46.4 -> 57.1Improved
use pre-trained multimodal model (MedViLL) instead of Fusion TransformerECG-QA baseline · ECG-QA test, S-VerifyEM accuracy: 72.1 -> 73.9Improved
remove ECG input (Blind Transformer)Fusion Transformer · ECG-QA test, S-VerifyEM accuracy: 72.1 -> 67.7Hurt
remove question input (Deaf Transformer)Fusion Transformer · ECG-QA test, CC-VerifyEM accuracy: 71.9 -> 59.5Hurt
use Verify instead of Query questionsM3AE · ECG-QA testEM accuracy: 41.0 -> 75.5Improved
replace best QA model with best upper-bound model (SE-WRN) for perceptionSingle-Verify · ECG-QA Single-Verifyaccuracy: 80.8 -> 83.1Improved
replace best QA model with MAX upper boundSingle-Verify · ECG-QA Single-VerifyAUROC: 0.808 -> 0.907Improved
replace trained QA model with zero-shot GPT-4 on SE-WRN textECG-QA · ECG-QA 10% test, S-ChooseEM accuracy: 58.2 -> 48.1Hurt
replace M3AE with SE-WRN + text-davinci-003ECG-QA · ECG-QA 10% test, S-VerifyEM accuracy: 76.0 -> 75.0Hurt
use gpt-4 instead of gpt-3.5-turboSE-WRN + LLM · ECG-QA 10% test, S-ChooseEM accuracy: 36.1 -> 48.1Improved
use gpt-4 vs majority baseline on CC-VerifySE-WRN + gpt-4 · ECG-QA 10% test, CC-VerifyEM accuracy: 58.4 -> 54.9Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy